STMicroelectronics Unveils Advanced Optical Interconnect Technologies for AI Clusters

Feb 21, 2025
STMicroelectronics has announced new silicon photonics and BiCMOS technologies to enhance optical interconnect performance in datacenters and AI clusters, according to a press release.

STMicroelectronics has announced new silicon photonics and BiCMOS technologies to enhance optical interconnect performance in datacenters and AI clusters, according to a press release. These advancements aim to support the increasing demands of AI computing by improving performance and energy efficiency.

The new technologies are designed to address the upcoming 800Gb/s and 1.6Tb/s optical interconnects. STMicroelectronics is collaborating with partners across the value chain to develop a roadmap for higher energy efficiency in pluggable optics, which are crucial for next-generation AI cluster GPU interconnects.

ST's proprietary silicon photonics technology allows for the integration of multiple complex components into a single chip, while the next-gen BiCMOS technology offers ultra high-speed and low power optical connectivity. These technologies will be manufactured on 300mm processes in Europe, providing an independent high-volume supply for optical module development.

The company plans to ramp up production of these technologies from the second half of 2025, aiming to become a key supplier in the datacenter and AI cluster market.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Silicon Brief or Daily AI Brief.

Also, consider following us on social media:

AI Chips & Datacenters AI Brief AI Brief (X)
Free newsletter

Silicon Brief

Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.